embedding-condensation
Measure layer-wise token embedding cosine similarity, assessing the severity of embedding condensation. Concept from [ICML 2026] Dispersion loss counteracts embedding condensation and improves generalization in small language models.
embedding-condensation has been downloaded 5,895 times in total on PyPI, including 268 in the last 30 days. The latest version is 2.0, released May 17, 2026.
Version2.0
Downloads
5.89k
LicenseNon-Commercial License — see ../LICENSE.md in the LM-Dispersion repository.
AuthorChen Liu
Downloads
Weekly, last 90d.
Includes CI traffic.
VersionsTotal2.*1.*
Range
View
Granularity
Group by
CI traffic
Stack: OffCI: Included3 / 5 series
Selected total1.18k
5.9kAll-time
268Last 30 days
1Last 24 h
<0.01/sPer second
Sponsored
Sponsorships keep pepy free to read
Version distribution
Share of downloads by released version. Computed over the last quarter.
- 0143.5%
2.0
279 downloadsDownloads27943.5% - 0220.6%
1.1
132 downloadsDownloads13220.6% - 0319.3%
1.0.0
124 downloadsDownloads12419.3% - 0416.5%
0.1.0
106 downloadsDownloads10616.5%
Guess the next day
Thirteen recent days of embedding-condensation downloads. Drag the green handle on the right to guess where day fourteen lands.